Do you need statistics to understand your data?

Friday, October 7th, 2011

The joys of having to discuss another statistical question: when I was thinking about this question I was like what data? Who was more likely to win the grand final St Helens or Leeds Rhinos? Which mascara would make my eyes lashes longer or which aftershave is more likely to help the boys get laid? We receive data every minute of every day with any action I take. However in these situations I don’t actually have time to make a statistical analysis on what best course of action to take except when I’m in the bookies. So in everyday life we don’t really need stats to understand what is best to do and somehow we seem to cope and my eyelashes still look good.       

However, when we spend hours doing our research and collecting our data, stats are a good tool to use when trying to understand our data and can be used to represent our findings in a single line of a statistical statement and with the help of the beautiful invention of SPSS which does this all for us we can easily analyze our data.  With the help of graphs and tables we can discuss our findings. If one relies purely on stats, they may miss something crucial and glaringly obvious about the data being analyzed. An experimenter can analyze the data in the wrong way or choose accidentally not to take into account something that has controlled the course of the data because they may not believe its important data with data such as age or gender.

We do stats to help us to understand our data in a manageable concept however we do need to take into account other aspects of our research to gain a representative conclusion of our study.

Advertisements

12 comments on “Do you need statistics to understand your data?

  1. I found your blog to be great! It was a perfect length and made clear points and maintained a sense of humour throughout. Frankly I agree with you competly, and although there may be other points to be made on this topic I feel you have pointed out the most important and relevant of these which is that we need stats, but not necessarily rely on them completely. Good job!

  2. psyalo says:

    Good blog. But don’t you think it’s important to understand how SPSS works, and what each button does before depending on it too heavily? For example the program does not take into account outlying data, and this could mess up your results and you would then come up with a false result.

    • thought3 says:

      I do not totally agree with your reasoning here. First of all, if you are going to use something without knowing or learning how to use it, you are only setting yourself up to fail to begin with. I do not think that SPSS is there to remove outliers out atall. It is your job as a researcher to see your outliers through your raw data and then with the easiest way (manually), remove them. The programme is there to assist you in organising your data that it can be easily readable and be able to make inferences. By you relying on a programme or software to detect or remove outliers for you then you are just being a lazy researrcher.

  3. lrowlands1 says:

    I agree with how stats can cause us to miss something crucial. They are important, like you said, to see the simplified picture of the data and what is shows, but most of the time it aids us in answering our question, to see was there a difference or not, However, data from a research can sometimes tell us so much more. So in order to gain a greater understanding we need both statistics and we need to look in greater depth at these numbers and what they mean and that can sometimes be done better without the use of statistics. I do disagree that statistics aren’t usually used in real life situations. Statistics are all around us, without us realising it. Think of consumerism for example, the research and statistical analysis that we know nothing about has ended up in morrisons own soup being on sale during freshers week when students can’t be bothered cooking, are usually hungover or is suffering from freshers flu. Crap example, but you get my drift?

    x

  4. lrowlands1 says:

    I agree with how stats can cause us to miss something crucial. They are important, like you said, to see the simplified picture of the data and what is shows. However, data from a research can sometimes tell us so much more. So in order to gain a greater understanding we need both statistics and we need to look in greater depth at these numbers and what they mean and that can sometimes be done better without the use of statistics. I do disagree that statistics aren’t usually used in real life situations. Statistics are all around us, without us realising it. Think of consumerism for example, the research and statistical analysis that we know nothing about has ended up in morrisons own soup being on sale during freshers week when students can’t be bothered cooking, are usually hungover or is suffering from freshers flu. Crap example, but you get my drift?

    x

  5. dnf24 says:

    Good blog, got right to the point and was not boring to read in the slightest. Although you said that you don’t need statistics in everyday life, I believe that statistics are there, in places where you would not expect. They are not necessary but I think a mention should go to where they are used in everyday life, such as in your university grades, insurance prices and every time you pick up a food item with percentages of ingredients. I also agree with lrowlands1 when they mentioned how statistics can tell us more than just looking at raw data (Peter D. Mlynek) Statistics can show you the average amount of time it took participants to find the x on the screen or how many times someone did an extra response. This can help us to find outliers easily, something which can affect our data immensely but may take a lot longer to find in raw data.

  6. Cashy says:

    Yeah again very good blog, but I do agree with what psyalo said about SPSS. It is crucial to know what the buttons on SPSS mean! Otherwise your wasting your time. And as mentioned above as an example, SPSS will not take outlying data into account, guaranteed to screw up results. Also on your conclusion you mentioned how we need to take in to account other aspects of our research methods when gaining a representative result… such as?! For example, past research can be a big help.You also mentioned how if we use stats too dependently we may miss out on crucial figures etc, same can be said if we don’t rely on stats enough surely. Its about developing a balance between using stats and other means.

  7. thought3 says:

    Going back to how you opened up your blog, for me in everyday life that’s when I think it may be more important to have the statistical way of analysing and interpreting. You would definitely look like a mad man with your analysis all the time but it could save money and your time. For instance you said something about betting, right now I am just trying to get my head round this stats thing but if I got to the place where I did not need to refer to a book all the time, I think it would be one of the best achievements because in the sense of betting you would think how is it that the house wins most of the money all the time??? what do they do and how can one go around it and win??? But in most cases you are right, there are situations where we definitely need both stats and other forms analysing. But stats is in the little things and its the little things that no one pays attention to that have the biggest effect.

  8. Joe Butler says:

    Hi there – I just want to check if you have made any comments for the last week? If you have, could you please create a blogpost directing me to where they are.

    Thanks,
    Joe

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: